1 /* $NetBSD: citrus_utf8.c,v 1.16 2007/03/06 16:13:58 tnozaki Exp $ */
4 * Copyright (c)2002 Citrus Project,
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
10 * 1. Redistributions of source code must retain the above copyright
11 * notice, this list of conditions and the following disclaimer.
12 * 2. Redistributions in binary form must reproduce the above copyright
13 * notice, this list of conditions and the following disclaimer in the
14 * documentation and/or other materials provided with the distribution.
16 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
17 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
18 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
19 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
20 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
21 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
22 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
23 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
24 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
25 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
31 * The Regents of the University of California. All rights reserved.
33 * This code is derived from software contributed to Berkeley by
34 * Paul Borman at Krystal Technologies.
36 * Redistribution and use in source and binary forms, with or without
37 * modification, are permitted provided that the following conditions
39 * 1. Redistributions of source code must retain the above copyright
40 * notice, this list of conditions and the following disclaimer.
41 * 2. Redistributions in binary form must reproduce the above copyright
42 * notice, this list of conditions and the following disclaimer in the
43 * documentation and/or other materials provided with the distribution.
44 * 3. Neither the name of the University nor the names of its contributors
45 * may be used to endorse or promote products derived from this software
46 * without specific prior written permission.
48 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
49 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
50 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
51 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
52 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
53 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
54 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
55 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
56 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
57 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
61 #include <sys/types.h>
72 #include "citrus_namespace.h"
73 #include "citrus_types.h"
74 #include "citrus_module.h"
75 #include "citrus_ctype.h"
76 #include "citrus_stdenc.h"
77 #include "citrus_utf8.h"
80 /* ----------------------------------------------------------------------
81 * private stuffs used by templates
84 static int _UTF8_count_array[256];
85 static int const *_UTF8_count = NULL;
87 static const u_int32_t _UTF8_range[] = {
89 0x00000000, 0x00000080, 0x00000800, 0x00010000,
90 0x00200000, 0x04000000, 0x80000000,
102 _UTF8EncodingInfo ei;
104 /* for future multi-locale facility */
107 _UTF8State s_mbrtowc;
109 _UTF8State s_mbsrtowcs;
110 _UTF8State s_wcrtomb;
111 _UTF8State s_wcsrtombs;
116 #define _CEI_TO_EI(_cei_) (&(_cei_)->ei)
117 #define _CEI_TO_STATE(_ei_, _func_) (_ei_)->states.s_##_func_
119 #define _FUNCNAME(m) _citrus_UTF8_##m
120 #define _ENCODING_INFO _UTF8EncodingInfo
121 #define _CTYPE_INFO _UTF8CTypeInfo
122 #define _ENCODING_STATE _UTF8State
123 #define _ENCODING_MB_CUR_MAX(_ei_) 6
124 #define _ENCODING_IS_STATE_DEPENDENT 0
125 #define _STATE_NEEDS_EXPLICIT_INIT(_ps_) 0
129 _UTF8_init_count(void)
133 memset(_UTF8_count_array, 0, sizeof(_UTF8_count_array));
134 for (i = 0; i <= 0x7f; i++)
135 _UTF8_count_array[i] = 1;
136 for (i = 0xc0; i <= 0xdf; i++)
137 _UTF8_count_array[i] = 2;
138 for (i = 0xe0; i <= 0xef; i++)
139 _UTF8_count_array[i] = 3;
140 for (i = 0xf0; i <= 0xf7; i++)
141 _UTF8_count_array[i] = 4;
142 for (i = 0xf8; i <= 0xfb; i++)
143 _UTF8_count_array[i] = 5;
144 for (i = 0xfc; i <= 0xfd; i++)
145 _UTF8_count_array[i] = 6;
146 _UTF8_count = _UTF8_count_array;
151 _UTF8_findlen(wchar_t v)
156 c = (u_int32_t)v; /*XXX*/
157 for (i = 1; i < sizeof(_UTF8_range) / sizeof(_UTF8_range[0]) - 1; i++)
158 if (c >= _UTF8_range[i] && c < _UTF8_range[i + 1])
161 return -1; /*out of range*/
165 _UTF8_surrogate(wchar_t wc)
167 return wc >= 0xd800 && wc <= 0xdfff;
172 _citrus_UTF8_init_state(_UTF8EncodingInfo *ei, _UTF8State *s)
179 _citrus_UTF8_pack_state(_UTF8EncodingInfo *ei, void *pspriv,
182 memcpy(pspriv, (const void *)s, sizeof(*s));
187 _citrus_UTF8_unpack_state(_UTF8EncodingInfo *ei, _UTF8State *s,
190 memcpy((void *)s, pspriv, sizeof(*s));
194 _citrus_UTF8_mbrtowc_priv(_UTF8EncodingInfo *ei, wchar_t *pwc, const char **s,
195 size_t n, _UTF8State *psenc, size_t *nresult)
202 _DIAGASSERT(nresult != NULL);
203 _DIAGASSERT(s != NULL);
204 _DIAGASSERT(psenc != NULL);
209 _citrus_UTF8_init_state(ei, psenc);
210 *nresult = 0; /* state independent */
214 /* make sure we have the first byte in the buffer */
215 if (psenc->chlen == 0) {
218 psenc->ch[psenc->chlen++] = *s0++;
221 c = _UTF8_count[psenc->ch[0] & 0xff];
222 if (c < 1 || c < psenc->chlen)
226 wchar = psenc->ch[0] & 0xff;
228 while (psenc->chlen < c) {
231 psenc->ch[psenc->chlen++] = *s0++;
233 wchar = psenc->ch[0] & (0x7f >> c);
234 for (i = 1; i < c; i++) {
235 if ((psenc->ch[i] & 0xc0) != 0x80)
238 wchar |= (psenc->ch[i] & 0x3f);
240 if (_UTF8_surrogate(wchar) || _UTF8_findlen(wchar) != c)
245 *nresult = (wchar == 0) ? 0 : s0 - *s;
252 *nresult = (size_t)-1;
257 *nresult = (size_t)-2;
262 _citrus_UTF8_wcrtomb_priv(_UTF8EncodingInfo *ei, char *s, size_t n, wchar_t wc,
263 _UTF8State *psenc, size_t *nresult)
268 _DIAGASSERT(nresult != NULL);
269 _DIAGASSERT(s != NULL);
271 if (_UTF8_surrogate(wc)) {
275 cnt = _UTF8_findlen(wc);
276 if (cnt <= 0 || cnt > 6) {
277 /* invalid UCS4 value */
282 /* bound check failure */
289 for (i = cnt - 1; i > 0; i--) {
290 s[i] = 0x80 | (c & 0x3f);
297 s[0] &= (0x7f >> cnt);
298 s[0] |= ((0xff00 >> cnt) & 0xff);
302 *nresult = (size_t)cnt;
306 *nresult = (size_t)-1;
312 _citrus_UTF8_stdenc_wctocs(_UTF8EncodingInfo * __restrict ei,
313 _csid_t * __restrict csid,
314 _index_t * __restrict idx,
318 _DIAGASSERT(csid != NULL && idx != NULL);
321 *idx = (_citrus_index_t)wc;
328 _citrus_UTF8_stdenc_cstowc(_UTF8EncodingInfo * __restrict ei,
329 wchar_t * __restrict wc,
330 _csid_t csid, _index_t idx)
333 _DIAGASSERT(wc != NULL);
345 _citrus_UTF8_stdenc_get_state_desc_generic(_UTF8EncodingInfo * __restrict ei,
346 _UTF8State * __restrict psenc,
347 int * __restrict rstate)
350 if (psenc->chlen == 0)
351 *rstate = _STDENC_SDGEN_INITIAL;
353 *rstate = _STDENC_SDGEN_INCOMPLETE_CHAR;
360 _citrus_UTF8_encoding_module_init(_UTF8EncodingInfo * __restrict ei,
361 const void * __restrict var, size_t lenvar)
370 _citrus_UTF8_encoding_module_uninit(_UTF8EncodingInfo *ei)
375 /* ----------------------------------------------------------------------
376 * public interface for ctype
379 _CITRUS_CTYPE_DECLS(UTF8);
380 _CITRUS_CTYPE_DEF_OPS(UTF8);
382 #include "citrus_ctype_template.h"
384 /* ----------------------------------------------------------------------
385 * public interface for stdenc
388 _CITRUS_STDENC_DECLS(UTF8);
389 _CITRUS_STDENC_DEF_OPS(UTF8);
391 #include "citrus_stdenc_template.h"